© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 12027
    Prev
    Next
    INDEX
    Explanations

    references to death and the death penalty

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/4-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    279
    +0.13
    0.7%
    106
    +0.11
    0.6%
    110
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    189
    +0.13
    0.03
    279
    +0.11
    0.03
    164
    +0.10
    0.03
    Negative Logits
    kered
    -1.98
    {}{
    -1.84
    tern
    -1.73
    ators
    -1.67
    yx
    -1.64
    cha
    -1.59
    appa
    -1.59
    ked
    -1.58
    quires
    -1.56
    uki
    -1.54
    POSITIVE LOGITS
     penalty
    2.53
     threats
    2.29
     toll
    2.17
     sentence
    2.07
     threat
    1.98
     certificates
    1.97
     sentences
    1.97
    match
    1.96
    blow
    1.96
     spiral
    1.94
    Activations Density 0.231%

    No Known Activations